U.S. patent number 8,209,112 [Application Number 11/992,498] was granted by the patent office on 2012-06-26 for method and device for operating an internal combustion engine.
This patent grant is currently assigned to Robert Bosch GmbH. Invention is credited to Markus Deissler, Michael Drung, Dirk Hartmann, Nicolas Ide, Georg Mallebrein, Lutz Reuschenbach, Andreas Roth, Frank Schiller.
United States Patent |
8,209,112 |
Hartmann , et al. |
June 26, 2012 |
Method and device for operating an internal combustion engine
Abstract
A method and a device are provided for operating an internal
combustion engine which allow for an improved diagnosis of the
valve mechanism of cylinders of the internal combustion engine. For
this purpose, a variable characteristic of a suction performance of
a cylinder of the internal combustion engine is ascertained. The
variable characteristic of the suction performance is ascertained
as a function of the mass flow flowing into an intake manifold of
the internal combustion engine and of a change of the intake
manifold pressure during an intake phase of the cylinder.
Inventors: |
Hartmann; Dirk (Stuttgart,
DE), Mallebrein; Georg (Korntal-Muenchingen,
DE), Ide; Nicolas (Ludwigsburg, DE), Roth;
Andreas (Muehlacker-Lomersheim, DE), Reuschenbach;
Lutz (Stuttgart, DE), Schiller; Frank (Tamm,
DE), Deissler; Markus (Neckarsulm, DE),
Drung; Michael (Muehlacker, DE) |
Assignee: |
Robert Bosch GmbH (Stuttgart,
DE)
|
Family
ID: |
37575117 |
Appl.
No.: |
11/992,498 |
Filed: |
September 29, 2006 |
PCT
Filed: |
September 29, 2006 |
PCT No.: |
PCT/EP2006/066930 |
371(c)(1),(2),(4) Date: |
October 14, 2009 |
PCT
Pub. No.: |
WO2007/036576 |
PCT
Pub. Date: |
April 05, 2007 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20100036580 A1 |
Feb 11, 2010 |
|
Foreign Application Priority Data
|
|
|
|
|
Sep 30, 2005 [DE] |
|
|
10 2005 047 446 |
|
Current U.S.
Class: |
701/114; 702/187;
701/103; 123/406.45; 123/347 |
Current CPC
Class: |
F02D
41/187 (20130101); F02D 35/025 (20130101); F02D
41/22 (20130101); F02D 2200/0406 (20130101); F02D
2200/0411 (20130101); F02D 2200/0402 (20130101); F02D
2200/0404 (20130101); Y02T 10/40 (20130101); F02D
41/0002 (20130101); F02D 41/1402 (20130101) |
Current International
Class: |
G06F
19/00 (20110101); G06F 17/40 (20060101); F02D
45/00 (20060101) |
Field of
Search: |
;701/102-105,110,114,115
;123/406.45,347,348 ;73/114.26,114.27,114.32,114.77,114.79
;702/182,183,187 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
4325902 |
|
Feb 1995 |
|
DE |
|
10064651 |
|
Jul 2002 |
|
DE |
|
0 399 829 |
|
Nov 1990 |
|
EP |
|
1247967 |
|
Oct 2002 |
|
EP |
|
1 643 101 |
|
Apr 2006 |
|
EP |
|
2002-227666 |
|
Aug 2002 |
|
JP |
|
2004-36610 |
|
Feb 2004 |
|
JP |
|
Other References
International Search Report, PCT International Patent Application
No. PCT/EP2006/066930, dated Jan. 19, 2007. cited by other.
|
Primary Examiner: Wolfe, Jr.; Willis
Assistant Examiner: Hoang; Johnny
Attorney, Agent or Firm: Kenyon & Kenyon LLP
Claims
What is claimed is:
1. A method for operating an internal combustion engine,
comprising: ascertaining a variable characteristic of a suction
performance of a cylinder of the internal combustion engine as a
function of a mass flow flowing into an intake manifold of the
internal combustion engine; ascertaining a change of an intake
manifold pressure during an intake phase of the cylinder; and
comparing a value resulting for the characteristic variable to a
setpoint value; and one of diagnosing, as a function of the result
of the comparison, a functioning of a suction of the cylinder, or
controlling at least one gas exchange valve of the cylinder to make
the variable characteristic of the suction performance follow the
setpoint value.
2. The method as recited in claim 1, wherein the intake manifold
pressure is ascertained by sampling using a first specified
sampling rate in a first time or crank angle interval specified in
its length and position.
3. The method as recited in claim 2, wherein multiple values for
the intake manifold pressure of the first specified time or crank
angle interval are averaged in a weighted manner.
4. The method as recited in claim 2, wherein the mass flow flowing
into the intake manifold is ascertained by sampling using a second
specified sampling rate, in a second time or crank angle interval
specified in terms of its length and position.
5. The method as recited in claim 2, wherein the mass flow flowing
into the intake manifold is modeled or calculated from performance
characteristics of the internal combustion engine in a specified
computing grid in a second time or crank angle interval specified
in terms of its length and position.
6. The method as recited in claim 4, wherein multiple values for
the mass flow of the second specified time or crank angle interval
are averaged in a weighted manner.
7. The method as recited in claim 4, wherein the second time or
crank angle interval is offset with respect to the first time or
crank angle interval by a specified time or crank angle span, by
one half of an ignition interval of two cylinders of the internal
combustion engine that are fired directly in succession in time at
least partly prior to the first time or crank angle interval.
8. The method as recited in claim 1, wherein the characteristic
variable is ascertained for multiple cylinders and values of
multiple cylinders resulting for the characteristic variable are
compared to one another and the functioning of the suction of the
cylinders is diagnosed as a function of the result of the
comparison.
9. The method as recited in claim 4, wherein as a first variable
characteristic of the suction performance a charge flowing off into
the cylinder is selected that is ascertained as a sum of a charge
flowing into an induction manifold of the last second specified
time or crank angle interval and a quotient of the difference
between the intake manifold pressures of the last two first
specified time or crank angle intervals and a constant.
10. The method as recited in claim 9, wherein a second variable
characteristic of the suction performance is ascertained as a
quotient of the first characteristic variable and a difference
between the intake manifold pressure and a partial pressure of the
last first time or crank angle interval that is a function of
residual gas in the cylinder.
11. The method as recited in claim 10, wherein a third variable
characteristic of the suction performance is formed as a quotient
of the second characteristic variable and an expected value.
12. The method as recited in claim 1, wherein the intake manifold
pressure is modeled by integrating a difference between a charge
flowing into an intake manifold and a charge flowing off into the
cylinder, the modeled intake manifold pressure is compared to a
measured intake manifold pressure and the variable characteristic
of the suction performance is selected using a charge exchange
model for determining the charge flowing off into the cylinder as a
function of the modeled intake manifold pressure in such a way that
the modeled intake manifold pressure is adapted to the measured
intake manifold pressure.
13. A device for operating an internal combustion engine,
comprising: an ascertainment unit adapted to ascertain a variable
characteristic of a suction performance of a cylinder of the
internal combustion engine, the ascertainment unit ascertaining the
variable characteristic of the suction performance as a function of
a mass flow flowing into an intake manifold of the internal
combustion engine and a change of an intake manifold pressure
during an intake phase of the cylinder; and a component adapted to
compare a value resulting for the characteristic variable to a
setpoint value, and, as a function of the result of the comparison,
diagnoses a functioning of a suction of the cylinder or makes the
variable characteristic of the suction performance follow the
setpoint value by controlling at least one gas exchange valve of
the cylinder.
Description
FIELD OF THE INVENTION
The present invention relates to a method and a device for
operating an internal combustion engine.
BACKGROUND INFORMATION
A method and a device for controlling a gas charge of a plurality
of cylinders in an internal combustion engine having a variable
valve control, in which a detection signal of a charge sensor is
sampled using a sampling rate, are described in German Patent No.
DE 10064651 A1. Furthermore, a detection interval is determined for
a cylinder. Within this detection interval, the sampling values are
added up to ascertain a sampling value sum. Furthermore, a number
of sampling values, which are within the first detection interval,
is counted for ascertaining a first count value. The air mass
filled into the first cylinder is then ascertained by forming a
quotient from the sampling value sum and the count value. When the
ratio between the intake manifold pressure and the ambient pressure
is greater than 0.8, gas charge deviations between the cylinders
are preferably detected using a hot film air mass sensor. When the
throttling is increased, i.e., when the ratio between the intake
manifold pressure and the ambient pressure is smaller than 0.8,
then the detection signal of the intake manifold pressure sensor is
preferably to be used for detecting gas charge deviations between
cylinders.
SUMMARY
An example method according to the present invention and an example
device according to the present invention for operating an internal
combustion engine may have the advantage that the variable
characteristic of the suction performance is ascertained as a
function of a mass flow flowing into an intake manifold of the
internal combustion engine and a change of the intake manifold
pressure is ascertained during an intake phase of the cylinder. In
this manner it is possible to implement a unified model for
ascertaining the variable characteristic of the suction performance
of the cylinder, which allows for the variable characteristic of
the suction performance in all possible throttle valve positions
and in all loads and engine speeds of the internal combustion
engine. In this context, the variable characteristic of the suction
performance may be determined precisely on the basis of a physical
model of the intake manifold and a physical model of the discharge
of the gas flow from the intake manifold into the cylinder having a
charge sensor system.
It may be particularly advantageous if the intake manifold pressure
is ascertained by sampling using a first specified sampling rate in
a first time or crank angle interval specified in its length and
position. In this manner, with a suitable choice of the length and
position of the specified time or crank angle interval, it is a
simple matter to ascertain the intake manifold pressure in
correlation to the respectively aspirating cylinder.
It may be furthermore advantageous if multiple values for the
intake manifold pressure of the first specified time or crank angle
interval are averaged, in particular in a weighted manner. In this
manner it is also sufficient to take samples for the intake
manifold pressure from the first specified time or crank angle
interval and to form from these a representative value for the
intake manifold pressure in the first specified time or crank angle
interval, it being possible, if the samples are weighted, to take
into account in a particularly simple manner a varying significance
of the samples for the characteristic curve of the intake manifold
pressure in the first specified time or crank angle interval.
Another advantage may be obtained if the mass flow flowing into the
intake manifold is ascertained, preferably by sampling using a
second specified sampling rate, in a second time or crank angle
interval specified in terms of its length and position. In this
manner, with a suitable choice of the second time or crank angle
interval specified in terms of its length and position, the mass
flow flowing into the intake manifold may be ascertained in a
particularly simple and reliable manner in correlation to the
respective currently aspirating cylinder.
The same is true in the event that the mass flow flowing into the
intake manifold is modeled or calculated from performance
characteristics of the internal combustion engine in a specified
computing grid in a second time or crank angle interval specified
in terms of its length and position.
It is furthermore advantageous if multiple values for the mass flow
of the second specified time or crank angle interval are averaged,
in particular in a weighted manner. In this manner it suffices to
take individual samples for the mass flow from the second specified
time or crank angle interval, which are representative for the
characteristic curve of the mass flow in the second specified time
or crank angle interval. In the event of a varying significance of
the individual samples for the characteristic curve of the mass
flow in the second specified time or crank angle interval, these
samples may also be averaged in a weighted manner.
Another advantage may be obtained if the second time or crank angle
interval is offset with respect to the first time or crank angle
interval by a specified time or crank angle span, preferably by one
half of an ignition interval of two cylinders of the internal
combustion engine that are fired directly in succession, in
particular in time at least partly prior to the first time or crank
angle interval. In this manner, the physical connection between the
mass flow flowing into the intake manifold and the intake manifold
pressure formed there for the respective currently aspirating
cylinder may be taken into account in an optimal manner for
ascertaining the variable characteristic of the suction performance
of the cylinder.
Another advantage may be obtained if a value resulting for the
characteristic variable is compared to a setpoint value and if as a
function of the result of the comparison the functioning of the
suction of the cylinder is diagnosed. In this manner it is possible
to check in a particularly simple manner with the aid of the
variable characteristic of the suction performance of the cylinder
whether or not the suction of the cylinder is error-free.
Another advantage may be obtained if the variable characteristic of
the suction performance is made to follow a setpoint value, in
particular by controlling at least one gas exchange valve of the
cylinder. In this manner, a cylinder-specific charge control may be
implemented in a particularly simple manner.
Another advantage may be obtained if the variable characteristic of
the suction performance is ascertained for multiple cylinders and
if values of multiple cylinders resulting for the characteristic
variable are compared to one another and if as a function of the
result of the comparison the functioning of the suction of the
cylinders is diagnosed. In this manner it is also possible to
detect the deviations in the suction performance of different
cylinders of the internal combustion engine and to diagnose an
underlying malfunction of the valve control.
A particularly simple ascertainment of the variable characteristic
of the suction performance is obtained if, as a first variable
characteristic of the suction performance, a charge flowing off
into the cylinder is selected that is ascertained as the sum of a
charge flowing into the intake manifold of the last second
specified time or crank angle interval and the quotient of the
difference between the intake manifold pressures of the last two
first specified time or crank angle intervals and a constant. In
this manner it is possible to ascertain the first variable
characteristic of the suction performance in a particularly simple
and reliable manner with the aid of an existing charge sensor
system.
On this basis, a second variable characteristic of the suction
performance may be ascertained as the quotient of the first
characteristic variable and a difference between the intake
manifold pressure and a partial pressure of the last first time or
crank angle interval that is a function of the residual gas in the
cylinder. In this manner, the second variable characteristic of the
suction performance is independent of the intake manifold pressure.
Thus, a multiplicative variable results for the cylinder of the
internal combustion engine, which characterizes the suction
performance.
Another advantage may be obtained if a third variable
characteristic of the suction performance is formed as the quotient
of the second characteristic variable and an expected value. This
again results in a multiplicative variable characterizing the
suction performance of the cylinder, which in addition is also
independent of the temperature of the intake air and the position
of the intake camshaft.
Another advantage may be obtained if the intake manifold pressure
is modeled by integrating the difference between a charge flowing
into the intake manifold and a charge flowing off into the
cylinder, if the thus modeled intake manifold pressure is compared
to a measured intake manifold pressure and if the utilized variable
characteristic of the suction performance is selected with the aid
of a charge exchange model for determining the charge flowing off
into the cylinder as a function of the modeled intake manifold
pressure in such a way that the modeled intake manifold pressure is
adapted to the measured intake manifold pressure. In this manner it
is possible to implement an adaptation of the variable
characteristic of the suction performance of the cylinder.
BRIEF DESCRIPTION OF THE DRAWINGS
An exemplary embodiment of the present invention is shown in the
drawings and explained in greater detail below.
FIG. 1 shows a block diagram of an internal combustion engine.
FIG. 2a) shows a diagram of a valve lift plotted against the crank
angle.
FIG. 2b) shows a diagram of an intake manifold pressure plotted
against the crank angle.
FIG. 2c) shows a diagram of a mass flow flowing into the intake
manifold plotted against the crank angle.
FIG. 3 shows a functional diagram for explaining the device
according to the present invention and the method according to the
present invention.
FIG. 4 shows a flow chart for an exemplary sequence of the method
according to the present invention.
DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS
In FIG. 1, 1 designates an internal combustion engine, which takes
the form of an Otto engine for example. Internal combustion engine
1 drives a motor vehicle for example. In the example shown in FIG.
1, it includes four cylinders 5, 10, 15, 20, of which a first
cylinder 5 is shown in exemplary fashion. Air is supplied to first
cylinder 5 via an air supply 45, which downstream of a throttle
valve 50 in air supply 45 turns into an intake manifold 25, and via
an intake valve 60. Furthermore, fuel is injected into intake
manifold 25 or directly into cylinder 5. The air/fuel mixture
present in cylinder 5 is ignited by a spark plug, which is likewise
not shown in FIG. 1. The subsequent combustion process drives a
piston (not shown in FIG. 1) of first cylinder 5, which in turn
drives a crankshaft of internal combustion engine 1. The exhaust
gas produced in the combustion of the air/fuel mixture is
discharged via an exhaust valve 65 of first cylinder 5 into an
exhaust branch 75 of internal combustion engine 1. The opening and
closing times of intake valve 60 and of exhaust valve 65 are
controlled either by a common camshaft or by a separate intake
camshaft and a separate exhaust camshaft or, as shown in FIG. 1,
directly by an engine control unit 35 using a variable valve
control. A crank angle sensor 70 situated in the region of
cylinders 5, 10, 15, 20 ascertains the current crank angle of
internal combustion engine 1 and relays it to engine control unit
35. Downstream from throttle valve 50, an intake manifold pressure
sensor 55 is situated in intake manifold 25, which measures the
current value of the intake manifold pressure and relays it to
engine control unit 35. In a manner known to one skilled in the
art, throttle valve 50 is controlled in its position by engine
control unit 35 as a function, for example, of a driver input or an
external control system such as, for example, a traction control
system, an antilock system, a vehicle speed controller, a driving
dynamics controller or the like and returns to engine control unit
35 a position feedback regarding the current position of throttle
valve 50, for example with the aid of a potentiometer. An air mass
meter 80, for example in the form of a hot film air mass meter, is
situated in air supply 45, which measures the air mass flow mszu
flowing into intake manifold 25 and relays the measured value to
engine control unit 35. The position value reported back from
throttle valve 50 exist for example in the form of the throttle
valve angle .alpha.. The intake manifold pressure ascertained by
intake manifold pressure sensor 55 is indicated in FIG. 1 by ps. In
FIG. 1, the crank angle value supplied by crank angle sensor 70 is
indicated by KW, and the engine speed derived from it by
differentiation is indicated by nmot. The other cylinders 10, 15,
20 operate in the manner described with respect to first cylinder
5.
The example method and the example device described in the
following allow for ascertaining a variable characteristic of the
suction performance of the individual cylinders 5, 10, 15,
connected to common intake manifold 25. Suction performance is here
understood as a variable typical of the charge exchange process,
which is characteristic of the fresh gas or air mass aspirated in
the intake stroke of the respective cylinder 5, 10, 15, 20. The
suction performance is influenced by the following variables: the
displacement of the respective cylinder 5, 10, 15, 20 the
temperature of the gas flowing into the respective cylinder 5, 10,
15, 20 the phase angle and the lift of the valve lifting curve of
the intake valve or intake valves of the respective cylinder 5, 10,
15, 20 and the tightness of the respective cylinder 5, 10, 15, 20,
in particular with respect to the intake and exhaust valves and the
piston rings.
The residual gas mass present in the respective cylinder 5, 10, 15,
20, which is characterized for example by the partial pressure
point of the residual gas, influences the suction performance as
well. In this example, the residual gas mass and thus the partial
pressure point of the residual gas is definitely specified,
however, so that for each cylinder 5, 10, 15, 20 it is necessary to
detect only one factor and not additionally an offset.
Ascertaining suction performance varying with respect to individual
cylinders is of particular interest for internal combustion engines
having more than one cylinder. This is required, among other
things, also for the diagnosis of valve control systems that switch
the lift of intake valves or even switch off entire cylinders. If
internal combustion engine 1 has a suitable continuous control
mechanism for influencing the suction performance in a
cylinder-specific manner, in particular for controlling the intake
and exhaust valves in a cylinder-specific manner, then it is also
possible to implement a cylinder-specific charge control by
ascertaining the variable characteristic of the suction performance
of the respective cylinder 5, 10, 15, 20.
For this purpose, according to the present invention, the variable
characteristic of the suction performance of the respective
cylinder 5, 10, 15, 20 is ascertained as a function of the gas mass
flow flowing into intake manifold 25 and the characteristic curve
of the intake manifold pressure in common intake manifold 25. For
this purpose, it must be possible to correlate individual cylinders
5, 10, 15, 20 and the ascertained values characteristic of the
suction performance.
According to the present invention, an intake manifold model and a
charge exchange model are used for ascertaining the value
characteristic of the suction performance and in particular its
cylinder-specific variations using the existing charge sensor
system described above, that is, air mass meter 80 and intake
manifold pressure sensor 55. The use of intake manifold pressure
sensor 55 is a prerequisite for the method according to the present
invention and the device according to the present invention. Air
mass meter 80 is advantageously provided, but its existence is not
essential. As an alternative to the main load signal mszu of air
mass meter 80 in the form of the air mass flow flowing to intake
manifold 25, it is also possible to evaluate as a secondary load
signal the signal .alpha. of the throttle valve angle sensor
developed in this example as a potentiometer, which is shown in
FIG. 1 by reference numeral 51. For this purpose, the air mass flow
mszu supplied to intake manifold 25 is calculated in a conventional
manner from throttle valve angle .alpha. and other performance
characteristics of internal combustion engine 1 such as, for
example, temperature T upstream from throttle valve 50 and the
pressure ratio across throttle valve 50. The pressure ratio across
throttle valve 50 is obtained as ps/pu, where pu is the ambient
pressure. The corresponding performance quantities T,pu may be
either detected by a suitable sensor system or may be modeled from
other performance characteristics of internal combustion engine 1
in a conventional manner.
Thus, for example, an intake air temperature sensor may be provided
in air supply 45 upstream from throttle valve 50 for ascertaining
the gas temperature upstream from throttle valve 50. Furthermore,
an ambient pressure sensor may likewise be provided upstream from
throttle valve 50 for ascertaining the ambient pressure in air
supply 45.
The measured intake air temperature is supplied to engine control
unit 35. The measured ambient pressure is likewise supplied to
engine control unit 35. Engine control unit 35 then ascertains the
air or gas mass flow mszu supplied to intake manifold 25 from
throttle valve angle .alpha., pressure ratio ps/pu across throttle
valve 50, and gas temperature T upstream from throttle valve 50 in
a conventional manner and as described.
Intake manifold pressure sensor 55 samples the intake manifold
pressure at a first specified sampling rate in a first time or
crank angle interval specified in terms of its length and position.
Subsequently, the sampled intake manifold pressure values are
transformed into a rotational speed-synchronous computing grid. The
transformation occurs by averaging the sampled intake manifold
pressure values over the time or crank angle interval specified in
its relative position and length. The averaging is performed
preferably by summing up the sampled intake manifold pressure
values in a temporally specified grid, for example 1 ms, and by
dividing the thus obtained sum by the number of grids per first
specified time or crank angle interval. The correlation between the
time interval and the associated crank angle interval is produced
via the current engine speed nmot in a conventional manner. The
averaging is performed alternatively by summing up only individual
samples at an arbitrary position of the first specified time or
crank angle interval and by dividing the obtained sum by the number
of samples taken. For this purpose, the samples may be taken at
positions of the first specified time or crank angle interval that
are considered to be particularly representative for the time
characteristic of the intake manifold pressure in the first
specified time or crank angle interval. In order to take into
account a varying significance of individual samples when
averaging, these may also enter the average for the intake manifold
pressure in the associated first specified time or crank angle
interval in a weighted manner.
Accordingly, mass flow mszu flowing into intake manifold 25 may
also be ascertained by air mass meter 80 by sampling using a second
specified sampling rate in a second time or crank angle interval
specified in terms of its length and position. For this purpose,
the second specified sampling rate may be advantageously selected
in accordance with the first specified sampling rate. The first
specified sampling rate, however, may also be selected to be
different from the second specified sampling rate. Alternatively,
the mass flow flowing into intake manifold 26 is modeled or
calculated in the manner described in a specified computing grid,
for example in grids of 1 ms in the second time or crank angle
interval specified in its length and position from the performance
characteristics of throttle valve angle .alpha., temperature T
upstream from throttle valve 50, and pressure ratio ps/pu across
throttle valve 50. Independently of the manner in which the mass
flow mszu flowing into intake manifold 25 is determined from the
main load signal or the secondary load signal, the transformation
of the sampling or computational values for the mass flow mszu
flowing into intake manifold 25 into a rotational speed-synchronous
computing grid occurs as described in that the sampling or
computational values for air mass flow mszu are averaged over the
second specified time or crank angle interval. This averaging
occurs in turn by summing up the measured values in the described
computing grid of 1 ms for example and by dividing the formed sum
by the number of computing grids per second specified time or crank
angle interval. Alternatively, individual samples may again be
ascertained at arbitrary positions of the second specified time or
crank angle interval for the air mass flow mszu flowing into intake
manifold 25 and added up in order subsequently to form an average
value by division by the number of samples. For this purpose, the
samples may be taken advantageously in positions of the second
specified time or crank angle interval that are of greater
significance for the characteristic curve of air mass flow mszu in
the second specified time or crank angle interval, it also being
possible for the samples to be averaged in a variously weighted
manner depending on the significance of the selected positions for
taking the samples.
In a four-stroke engine, a value is selected as the reference value
for the length of the first or of the second specified crank angle
interval, which may be described by the following equation:
Phi_length=720.degree. KW/number of cylinders (1)
Phi_length corresponds to the length of the first or second
specified crank angle interval, KW signifies the crank angle, and
the number of cylinders is the number of cylinders of internal
combustion engine 1, the number of cylinders in the present example
being=4.
Shorter or longer first or second specified crank angle intervals
may be selected as well, however.
According to equation (1), the first or second specified crank
angle intervals decrease with an increasing number of cylinders.
The important point is that the first and the second specified
crank angle interval may be assigned to the intake phase of the
cylinder considered for ascertaining the value characteristic of
the suction performance.
For ascertaining the intake manifold pressure, the phase angle of
the first specified crank angle interval may be adapted as a
function of the installed position of intake manifold pressure
sensor 55 in intake manifold 25 and especially as a function of
engine speed nmot and other parameters such as, for example, the
average value of intake manifold pressure ps. A proven favorable
value for averaging intake manifold pressure ps is obtained when
the first specified crank angle interval lies approximately in the
middle around the time "intake closes" of the respective cylinder.
This situation is illustrated in FIG. 2a) and FIG. 2b). FIG. 2a)
shows the characteristic curve of valve lift VH of the intake valve
of one of cylinders 5, 10, 15, 20 of internal combustion engine 1
plotted against crank angle KW. The characteristic curve of the
valve lift is represented by a dash-dot line, which has for
comparison the characteristic curve of the air mass flow to the
respective cylinder superimposed as a solid line. The middle of
first specified crank angle interval 300 is set approximately at
crank angle KW, at which valve lift VH and with it the mass flow
flowing to the respective cylinder starting from its maximum value
again reaches the value 0. It begins at a first crank angle .phi.1
before "intake closes" and ends at a second crank angle .phi.2
after "intake closes". According to the characteristic curve of
intake manifold pressure ps against crank angle KW as shown in FIG.
2b), the value of intake manifold pressure ps in the middle of the
first specified crank angle interval corresponds approximately to
the average value of the characteristic curve of the intake
manifold pressure ps against crank angle KW in the first specified
crank angle interval.
The average value of the intake manifold pressure ps in the first
specified crank angle interval 300 is indicated in FIG. 2b) by
reference numeral 305.
For detecting the mass flow, the second specified crank angle
interval for averaging air mass flow mszu may deviate in its length
and phase from the first specified crank angle interval for
averaging intake manifold pressure ps. Since in the first specified
crank angle interval the induction of gas from intake manifold 25
is detected, the averaging of air mass flow mszu should ideally
occur in a second specified crank angle interval that is offset by
a specified crank angle span with respect to the first specified
crank angle interval. In this instance, the second specified crank
angle interval is advantageously shifted in the advance direction
with respect to the first specified crank angle interval,
preferably by one half of an ignition interval of two cylinders
that are fired in direct succession. The first specified crank
angle interval and the second specified crank angle interval may
also overlap each other or, alternatively, may not have a common
intersection.
Such a phase shift of the specified crank angle intervals for
averaging across intake manifold pressure ps and air mass flow mszu
takes into account the fact that air mass flow mszu into intake
manifold 25 during the second specified crank angle interval yields
as the final value the ascertained and in particular averaged
intake manifold pressure in the first specified crank angle
interval. The signals of intake manifold pressure ps and air mass
flow mszu thus averaged in the intake phase of the respective
cylinder over the corresponding crank angle intervals are clearly
associated with this cylinder.
FIG. 2c) shows the air mass flow mszu flowing into the intake
manifold plotted against crank angle KW. In this instance, second
specified crank angle interval 310 from a third crank angle .phi.3
to a fourth crank angle .phi.4 is shifted in the advance direction
by the described half of an ignition interval with respect to first
specified crank angle interval 300 and overlaps with first
specified crank angle interval 300. Thus second specified crank
angle interval 310 covers the characteristic curve of valve lift VH
at maximum valve lift, at which the associated aspirating cylinder
has the greatest suction performance, such that in second specified
crank angle interval 310 a sharp rise of air mass flow mszu against
crank angle KW is registered and its average over second specified
crank angle interval 310 is indicated by reference numeral 315.
As shown in FIG. 2c), second specified crank angle interval 310 is
selected in such a way that fourth crank angle .phi.4 lies
approximately in the middle of first specified crank angle interval
300. The selection of the two specified crank angle intervals 300,
310 is based on the assumption that between third crank angle
.phi.3 and second crank angle .phi.2 only the cylinder associated
with the valve lift curve shown in FIG. 2a) aspirates, while the
other cylinders of the internal combustion engine are not
aspirating.
Now the differential equation of intake manifold 25 may be set up
as follows:
.times..times..intg.d.times..times. ##EQU00001##
Equation (2) thus represents the intake manifold model. To
implement this integral equation (2) in engine control unit 35, it
may be executed by a simple computing rule in a synchronous
computing grid. The calculation in a synchronous computing grid,
i.e., once per aspiration of a cylinder, has the consequence that
the calculation is performed, not on mass flows, but on charges.
rlab denotes the fresh air charge in the combustion chamber of the
respective cylinder which results during the intake phase of the
cylinder when air mass flow msab flows into the combustion chamber
of the respective cylinder. msab in equation (2) is thus the air
mass flow that flows into the combustion chamber of the cylinder
during its intake phase. Volume.sub.intake manifold is the volume
of intake manifold 25 and density.sub.gas is the density of the
fresh gas contained in the intake manifold. rlzu in turn is the
fresh air charge that enters into intake manifold 25 during the
intake phase of the respective cylinder when air mass flow mszu
flows to intake manifold 25. rlzu is a standardized charge,
assuming values between 0 and 100%, and resulting from the
following equation:
##EQU00002##
In equation (3), KUMSRL is a constant, dependent on the number of
cylinders and the displacement, for converting between mass flow
and charge and may be applied in a manner known to one skilled in
the art on a test stand, for example, or may be calculated from the
displacement and the number of cylinders of the engine.
The integral equation (2) of intake manifold 25 may now be written
as a sum equation with charges in the synchronous computing grid:
ps(n)=ps(n-1)+K.sub.intake[rlzu(n)-rlab(n)] (4)
Equation (4) means that intake manifold pressure ps(n) ascertained
for computing grid n results from intake manifold pressure ps(n-1)
ascertained for computing grid n-1 plus the difference between
charge rlzu(n) flowing into intake manifold 25 in computing grid n
and charge rlab(n) flowing out of intake manifold 25 into the
respective cylinder in computing grid 1 multiplied by a constant
K.sub.intake. Constant K.sub.intake is a function of the volume of
the intake manifold and the temperature in the intake manifold and
may be applied in a manner known to one skilled in the art on a
test stand, for example, or may be obtained from the geometric
dimensions of the intake manifold. Computing grid n-1 for the
respective cylinder is earlier than computing grid n by exactly one
synchronous computing grid, that is, it lies in the intake phase of
the respective cylinder directly preceding the intake phase of this
cylinder associated with computing grid n. Solving equation (4) for
rlab(n) yields: rlab(n)=rlzu(n)+(ps(n-1)-ps(n))/K.sub.intake
(5)
In equation (5), ps(n-1) and ps(n) is respectively the measured
variable for the intake manifold pressure averaged over the first
specified crank angle interval. rlzu is the charge signal obtained
via equation (3) from the measured or modeled air mass signal mszu,
which represents an average value over the second specified crank
angle interval.
In particular in the case of pressure ratios across throttle valve
50 of ps/pu<0.8, air mass flow mszu and thus charge rlzu may be
ascertained with the aid of the secondary load signal, as
described, i.e., calculated from a throttle valve model, throttle
valve position a, intake manifold pressure ps, ambient pressure pu
and temperature T upstream from throttle valve 50 then being taken
into account in the manner described.
Charge rlab(n) aspirated into the respective cylinder in accordance
with equation (5) represents a first variable characteristic of the
suction performance of the respective cylinder. Generally, however,
the variable rlab thus calculated is not yet the desired target
variable. Rather, in general, a variable is to be calculated, which
the suction performance of the respective cylinder be independent
of the intake manifold pressure and optionally also independent of
the temperature and the setpoint position of the camshaft. For this
reason, in addition to the intake manifold model in accordance with
equation (2), a charge exchange model is required as well. The
charge exchange model describes the fresh air charge rlab inducted
into the respective cylinder as a function of intake manifold
pressure ps. As already described, there are also additive portions
in the charge exchange model. These are combined by partial
pressure pbrint of the residual gas. This additive portion,
however, should not be calculated, but rather specified in a fixed
manner. Thus, for each cylinder 5, 10, 15, 20, only a
multiplicative variable, i.e., a factor describing the suction
performance of the respective cylinder, needs to be determined. The
charge exchange equation of the charge exchange model solved for
the factor of the conversion from pressure into charge reads as
follows: fupsrl(n)=rlab(n)/[ps(n)-pbrint(n)] (6)
Each of the variables specified in equation (6) again represents an
average value over synchronous computing grid n or the
corresponding first or second crank angle interval. In this
instance, factor fupsrl(n) for converting pressure into charge
represents a second value characteristic of the suction performance
of the respective cylinder.
In order to obtain an independence from temperature T of the intake
air, factor fupsrl(n) for converting pressure into charge may
itself in turn be related to an expected value fupsrlsetpoint(n):
fupsrl(n)=fupsrlsetpoint(n) factor.sub.fupstl (7)
Factor factor.sub.fupstl from equation (7), which is independent of
intake manifold pressure and temperature, represents a third
variable characteristic of the suction performance of the
respective cylinder. The factual situation represented in equations
(5) through (7) represents the central idea of the present
invention:
A value characteristic of the cylinder-specific suction performance
may be obtained from values of the current intake manifold pressure
and the intake manifold pressure in the previous synchronous
computing grid averaged over the first specified crank angle
interval and from an air mass flow into intake manifold 25 averaged
over the second specified crank angle interval, that is, as a
function of the air mass flow flowing into intake manifold 25 of
the internal combustion engine in the current intake phase and a
change of the intake manifold pressure in the previous two
consecutive intake phases of the respective cylinder. The method
according to the present invention is represented in an exemplary
manner in FIG. 4 in the form of a flow chart. Following the start
of the program, at a first program point 200 during the first
specified crank angle interval, engine control unit 35 detects the
sampling values of intake manifold pressure sensor 55 and detects
the sampling values of air mass meter 80 during the second
specified crank angle interval. In the case of ascertaining air
mass flow mszu from the secondary load signal, engine control unit
35 at program point 200 in the second specified crank angle
interval ascertains the respective computational values for air
mass flow mszu in the manner described. Subsequently, the program
branches to a program point 205.
At program point 205, engine control unit 35 forms an average value
of the sampled values for the intake manifold pressure or the
obtained samples for the intake manifold pressure in the first
specified crank angle interval. This average value is given by
variable ps(n). Furthermore, at program point 205, engine control
unit 35 ascertains the average value of the values for air mass
flow mszu detected or calculated in the second specified crank
angle interval or of the samples for air mass flow mszu formed in
this second specified crank angle interval such that with the aid
of equation (3) and the average value for engine speed nmot in the
second specified crank angle interval and the applied constant
KUMSRL the variable rlzu(n) for the charge supplied to intake
manifold 25 is obtained. At program point 205, engine control unit
35 furthermore ascertains value ps(n-1) from the previous computing
grid likewise as an average value of the intake manifold pressures
formed in the previous intake phase of the respective cylinder in
the respective first specified crank angle interval in the manner
described. Subsequently, the program branches to a program point
210.
At program point 210, engine control unit 35 calculates in
accordance with equation (5) the charge rlab(n) currently aspirated
by the respective cylinder as an average value in the manner
described. Subsequently, the first value characteristic of the
suction performance of the respective cylinders is obtained such
that the program may be terminated. Optionally, however, after
program point 210, the program branches to a program point 215.
At program point 215, in accordance with equation (6), factor
fupsrl(n) for converting pressure into charge is in turn calculated
in the form of an average value and represents the second variable
characteristic of the suction performance of the respective
cylinder. Subsequently, the program may be terminated.
Alternatively, however, the program may branch from program point
215 to a program point 220. At program point 220, in accordance
with equation (7), factor factor.sub.fupstl is calculated as an
average value and third variable characteristic of the suction
performance of the respective cylinder. Subsequently the program is
terminated.
As an alternative to calculating the respective variable
characteristic of the suction performance of the respective
cylinder with the aid of equations (5)-(7), the variable
characteristic of the cylinder-specific suction performance may
also be learned by adaptation. In this case, the intake manifold
and charge exchange model shown in FIG. 3 is implemented in the
engine control unit in terms of software and/or hardware and is
calculated during each intake process of a cylinder in a
synchronous computing grid. The intake manifold model and the
charge exchange model in this instance correspond exactly to
equations (5)-(7), the only difference being the fact that in the
adaptation the variable characteristic of the cylinder-specific
suction performance is adapted by matching a modeled intake
manifold pressure with the measured intake manifold pressure.
In FIG. 3, identical reference numerals denote the same elements as
in FIG. 1. For the exemplary embodiment as shown in FIG. 3 it is
assumed that air mass flow mszu is detected by air mass meter 80.
The current crank angle values KW detected by crank angle sensor 70
are supplied to a differentiating element 85 that ascertains the
time gradient of the crank angles detected by crank angle sensor 70
and supplies it as the engine speed nmot to a first multiplication
element 100, which is also supplied by a factor value memory 95
with factor KUMSRL. The product nmot*KUMSRL obtained at the output
of first multiplication element 100 is supplied as a divisor to a
division element 105, which is also supplied by air mass meter 80
with the measured air mass flow mszu as dividend. Thus, the
quotient rlzu=mszu/(nmot*KUMSRL) according to equation (3) is
obtained at the output of division element 105. From this quotient,
a first subtraction element 110 subtracts charge rlab flowing off
into the respective cylinder, which is formed by a charge exchange
model 30. Thus the difference rlzu-rlab is obtained at the output
of first subtraction element 110. This is supplied as input
variable to a first integrator 130, which represents the intake
manifold model. Thus a modeled value for intake manifold pressure
psmod is obtained at the output of first integrator 130. In a
second subtraction element 115, the intake manifold pressure ps
measured by intake manifold pressure sensor 55 is subsequently
subtracted from the modeled value psmod of the intake manifold
pressure. The resulting difference psmod-ps may be supplied via a
first controlled switch 140, depending on the position of the
switch, either to a second integrator 150, a third integrator 155,
a fourth integrator 160 or a fifth integrator 165. For this
purpose, second integrator 150 is associated with first cylinder 5,
third integrator 155 with second cylinder 10, fourth integrator 160
with third cylinder 15 and fifth integrator 165 with fourth
cylinder 20. The output signal of crank angle sensor 70 is also
supplied to an evaluation unit 135 of engine control unit 35, which
controls first controlled switch 140 to connect the output of first
subtraction element 115 to second integrator 150 in the intake
phase of first cylinder 5, to connect the output of first
subtraction element 115 to third integrator 155 during the intake
phase of second cylinder 10, to connect the output of second
subtraction element 115 to fourth integrator 160 during the intake
phase of third cylinder 15, and to connect the output of second
subtraction element 115 to fifth integrator 165 during the intake
phase of fourth cylinder 20. The outputs of integrators 150, 155,
160, 165 are able to be connected optionally to a second
multiplication element 125 of charge exchange model 30 via a second
controlled switch 160, which is switched by evaluation unit 135
synchronously with respect to first controlled switch 140. In this
instance, the output of second integrator 150 is connected to
second multiplication element 125 during the intake phase of first
cylinder 5, the output of third integrator 155 to second
multiplication element 125 during the intake phase of second
cylinder 10, the output of fourth integrator 160 to second
multiplication element 125 during the intake phase of third
cylinder 15, and the output of fifth integrator 165 is connected to
second multiplication element 125 in the intake phase of fourth
cylinder 20. In this instance, the output of integrators 150, 155,
160, 165 represents factor fupsrl for converting pressure into
charge. It is adapted by integrators 150, 155, 160, 165 in the
sense of minimizing the difference psmod-ps. In a third subtraction
element 120, the partial pressure pbrint of the residual gas from a
residual gas value memory 90 is subtracted from modeled intake
manifold pressure psmod on the output of first integrator 130. The
resulting difference psmod-pbrint on the output of third
subtraction element 120 is supplied to second multiplication
element 125 and is there multiplied by factor fupsrl for converting
pressure into charge such that at the output of second
multiplication element 125 charge rlab is obtained, which is
inducted into the respective cylinder and which is, as described,
supplied to first subtraction element 110. In this instance, third
subtraction element 120 and second multiplication element 125 form
charge exchange model 30. Integrators 150, 155, 160, 165 form an
ascertainment unit 40 for ascertaining a value characteristic of
the suction performance of the individual cylinders, that is, a
cylinder-specific suction performance in the form of factor fupsrl
for converting pressure into charge. With the adaptation of factor
fupsrl for converting pressure into charge, value rlab for the
charge inducted in the respective cylinder is also adapted as the
first value characteristic of the suction performance. As shown in
FIG. 3, residual gas value memory 90 and factor value memory 95 as
well as differentiating element 85 are situated outside of engine
control unit 35, but may optionally and independently of each other
also be implemented in engine control unit 35.
A refinement of the present invention optionally provides for the
value resulting for the utilized first, second or third variable
characteristic of the suction performance of a cylinder to be
compared to a setpoint value and for the functioning of the suction
of the respective cylinder to be diagnosed as a function of a
result of the comparison. This may occur at a program point 225
following program point 220, as shown in FIG. 4, program point 225
being represented by a dashed line. The setpoint value may be
applied on a test stand, for example. If the value resulting at
program point 225 for the utilized characteristic variable agrees
with the setpoint value within a specified tolerance range,
likewise applied on a test stand, for example, then an error-free
functioning of the suction of the respective cylinder is diagnosed,
otherwise an error in the suction of the respective cylinder is
diagnosed and an optical and/or acoustic warning message, if
necessary an emergency operation of the internal combustion engine,
or as a last resort a shutdown of the internal combustion engine is
initiated. According to another optional specific embodiment of the
present invention, the ascertained and utilized variable
characteristic of the suction performance of the respective
cylinder may also be used to perform a cylinder-specific charge
control in which the utilized variable characteristic of the
suction performance of the respective cylinder, which is
ascertained in the manner described, is made to follow a specified
setpoint value. In this case, the setpoint value may be
ascertained, for example, as a function of a driver input or a
requirement of an external control system such as for example a
traction control, an antilock system, a vehicle dynamics control, a
vehicle speed control or the like or may be applied as a fixed
value on a test stand, for example. The cylinder-specific charge
control may then be implemented for example by individually
controlling at least one gas exchange valve of the respective
cylinder, for example one or more intake valves and one or more
exhaust valves of the respective cylinder. This is possible
particularly in the case of a fully variable valve control in which
in this manner cylinder-specific fluctuations of the suction
performance are able to be detected and compensated or in which
cylinder-specific setpoint charges are deliberately controlled in
the manner described. A corresponding control step for the current
intake phase of the respective cylinder is carried out at program
point 225 as shown in FIG. 4. The program as shown in FIG. 4 is run
through for each intake phase of the currently considered cylinder.
Following program point 225, the program is terminated.
Another optional specific embodiment provides for the ascertained
utilized variable characteristic of the suction performance to be
ascertained for multiple cylinders, and for the values of multiple
cylinders resulting for the variable characteristic of the suction
performance to be compared to one another, and for the functioning
of the suction of the cylinders to be diagnosed as a function of
the result of the comparison. In this manner, a diagnosis of
undesired cylinder-specific variations between the suction
performances of individual cylinders may be diagnosed. A typical
application is the diagnosis of valve mechanisms for switching off
cylinders. Mistakenly switched-off or mistakenly active cylinders
are immediately detected. Furthermore, a clear cylinder allocation
may be made on the basis of the first and second crank angle
intervals selected as described. Additional applications are
conceivable in the case of valve mechanisms using lift changeover.
The varying suction performances when using varying lift curves in
different cylinders may be detected and may be compared to their
setpoint values for diagnostic purposes. Valve control systems
using phase control, where cylinder-specific variations may occur
due to the construction, may also be diagnosed in this manner. In
particular, variations in the suction performance of different
cylinders in electromagnetic or electrohydraulic fully variable
valve controls may be diagnosed in the manner described.
In the case of the described comparison of the values
characteristic of the suction performances of multiple cylinders,
variations in the suction performances of individual cylinders may
also be rectified in that the values characteristic of the suction
performances of the individual cylinders are controlled to match a
common setpoint value in order thus to achieve an equalization of
these cylinders with respect to the variable characteristic of the
suction performance.
The diagnosis of an error in the cylinder-specific suction
performance of a cylinder may be due to fact, for example, that the
piston rings are not longer sufficiently tight such that the charge
of the respective cylinder is reduced at low loads and engine
speeds by the fact that gas from the crankcase impairs the inflow
of fresh air via intake manifold 25. In this case, it is possible
to use the cylinder-specific diagnosis of the suction performance
to perform a compression diagnosis, in which the piston rings are
checked for sufficient tightness.
The described diagnoses may be performed, for example, at the end
of the assembly line following the manufacture of the internal
combustion engine or of the vehicle or in a workshop or even while
the internal combustion engine is in operation. In the process,
particularly at the end of the assembly line, it is possible to
check all possible variants of valve lift curves in their effect on
the charge of the individual cylinders in the manner described for
diagnostic purposes.
* * * * *